Clinical and research applications of magnetic resonance imaging continue to demand improvements in spatial and temporal resolution. Two of the most promising means of exceeding existing limits on spatial and temporal resolution include the use of high magnetic field strengths, on the one hand, and the use of radiofrequency (RF) coil arrays for accelerated parallel imaging, on the other. Particular synergies are expected for combinations of these approaches. Shortened acquisition times resulting from parallel imaging may be used to overcome some of the challenges associated with high-field studies, including susceptibility artifacts and specific absorption ratio (SAR) constraints. Meanwhile, increased spin polarization at high field strength results in increased signal to noise ratio (SNR), which can enable higher accelerations than would be possible at lower fields. Additional synergies have recently been predicted. Calculations of ultimate intrinsic SNR show marked improvements in achievable parallel imaging performance with increasing field strength, above and beyond the effects of increased spin polarization. These improvements have been traced to the reduced RF wavelength and improved RF focusing capability at high field strength. Access to the full benefits of this synergy between parallel imaging and high field strength will require changes in some of the traditional paradigms for coil array and RF system design. The broad goal of this proposal is to solve the theoretical and practical issues of RF design required in order to approach the computed optimal parallel imaging performance as a function of field strength. We propose to evaluate the efficacy of various decoupling strategies, to establish concrete benchmarks for practical coil sizes, and to build prototype many-element arrays capable of order-of-magnitude accelerations at 1.5T and 3T. In the course of the project, we will also establish the basic principles by which these designs may be extended to higher field strengths in order to yield still greater improvements in spatial and temporal resolution. Specific Aims of the proposed research are as follows: 1. Using 8-element test arrays, assess the impact of inter-element decoupling strategies upon baseline SNR, and implement the strategy that yields the best SNR. 2. Use new 32-receiver systems scheduled to be installed at Beth Israel Deaconess Medical Center to establish the smallest practical coil size for 32-element arrays as a function of field strength. 3. Based on the results from Specific Aims 1 and 2, construct 32-element arrays suitable for two-dimensional accelerations at 1.5T and at 3T, respectively, and perform quantitative comparisons of array performance. 4. Use two-dimensional acceleration in combination with these arrays to achieve order-of-magnitude increases in spatial and/or temporal resolution for a set of imaging sequences used commonly in cardiac, breast, body, and brain imaging applications. 5. Adapt target field methods from gradient coil design to establish robust parallel imaging array designs that approximate the computed optimum behavior at higher field strength.